Awesome

Vision Retrieval

Presentation

https://docs.google.com/presentation/d/1LY3KxUjuLAoCvKh9UyXtQupqiSTi_BFNUQ_Cdl6-o_g/edit#slide=id.g2f6ce2a35cc_0_103

TLDR

Setup

pip install modal
modal setup

Run

modal run vision_retrieval/infra.py

Deploy

modal deploy vision_retrieval/infra.py

Develop

modal shell vision_retrieval/infra.py

Orchestrate

pip install dagster dagster-webserver -U
dagster dev -f vision_retrieval/pipeline.py -p 3000 -h 0.0.0.0

References

• ColPali
• LanceDB
• ModalLab
• Dagster
• Beyond Text: The Rise of Vision-Driven Document Retrieval for RAG
• PDF Retrieval with Vision Language Models